Challenge

Over the last 15 years, web load balancers have been deployed to scale web applications. Web load balancers eventually became a requirement for any application networking build out. As web traffic demands increased, it was apparent that the database tier was trailing, becoming the weakest link in the application performance chain. In most cases, application bottlenecks are due to latent and inefficient database interaction.

“What is old is new again”

To solve this pain point, Heimdall Data developed a “SQL load balancer” to complement web balancers. SQL load balancers enable application owners to horizontally scale their data access layer. To best describe a SQL load balancer, I have created a comparison table below. The functions are strikingly similar to a web load balancer; the only difference being the protocol supported:

Feature

Web Load Balancer SQL Load Balancer
High Availability Failover

Security Firewall

Caching

Transformation

Connection Pooling

 

Design Intent

With small engineering teams and lowering budgets, organizations are challenged to continuously add application value while improving performance and scalability. Heimdall was developed for technical teams to focus on revenue-generating features offloading the burden of coding infrastructure optimization.

Heimdall is an off-the-shelf “Application Resiliency” platform for the database tier. Application owners can now be worry-free about performance, scale, high availability, and security at the data access layer. Why spend months or years developing application resiliency when you can install Heimdall in 5 minutes?

 “What makes Heimdall Data unique?”

  • Application-aware: most load balancing solutions drop the application connection upon a database failover or require DNS re-resolution. Heimdall Data maintains application uptime.
  • Self-healing: identify and resolve SQL bottlenecks (e.g. SQL caching, transformation).
  • Lightweight software solution: no hardware appliance required.
  • Easy to deploy: requires ZERO application code changes.
  • Fast to test, easy to use: don’t waste months developing a high availability solution for MySQL and PostgreSQL when you can get Heimdall configured and running in minutes.

In my next blog, I will discuss how Heimdall’s SQL load balancer is good for on-premise applications, but even better for the CLOUD.

Facebooktwittergoogle_plusredditpinterestlinkedinmail